From GIS to Remote Sensing
During the last few months I have been developing a new version of the Semi-Automatic Classification Plugin (SCP) for QGIS. This new SCP version 4.0 codename "Frascati" (dedicated to the ESA ESRIN located in Frascati (Italy), which is the European centre of Earth observation missions, and it is involved in the development of the Copernicus Sentinel satellites in particular for the acquisition, distribution and exploitation of data) will be officially released on February 16th 2015.
I have tried to add several new functions both to the user interface and the processing tools. In particular, a new SCP menu in located in the main menu bar of QGIS. One of the main achievements is the optimization of the Maximum Likelihood algorithm, which is considerably faster than before.
Another major change is that temporary ROIs are displayed as overlay, avoiding the loading of temporary shapefiles in QGIS. Also, new options allow for the automatic refresh of temporary ROIs changing ROI parameters.
A combo box allows for the rapid visualization of RGB color composites, which is useful for image interpretation.
New custom functions are available for splitting raster bands, converting classifications to shapefile, reclassifying classifications (also using Python expressions).
I have summarized the main changes in the following video.
Following the list of main changes:
-new SCP menu in the main menu bar
-enhanced the Maximum Likelihood algorithm which is now considerably faster
-temporary ROI are displayed as overlay (rubber band) with a + symbol on clicked pixel
-new option to hide and show the temporary ROIs
-in ROI dock new option for refreshing automatically the temporary ROI while changing ROI parameters
-in ROI dock new option for automatic plot of temporary ROI
-new option for displaying over the region growing cursor vegetation indices such as NDVI and EVI (the latter only for images converted to reflectance)
-new function for displaying the pixel spectral signature with right click on map using ROI pointer
-new combo box in SCP toolbar for quickly displaying RGB color composites of input image (and automatic creation of a virtual raster for band set)
-now the window spectral signature plot allows for the display of signature values in table
-new tool Split in Pre processing tab for splitting raster bands
-new tool Convert classification to vector in Post processing tab for creating a classification shapefile
-new tool Reclassification in Post processing tab for the easy reclassification from Class ID to Macroclass ID and Python expressions are allowed in the definition (e.g. raster < 3)
-enhanced function Clip multiple rasters
-new reset button for resetting the signature list file
-new button in the classification dock for merging spectral signatures obtaining a new one calculated as the average of signature values (without covariance matrix)
-now classes or macroclasses having ID = 0 are allowed and classified as 0 (unclassified) in the output which is useful for creating masks for classes
-in the signature list it is possible to set the color of multiple selected signatures at once (pressing shift and double click on a color) and check/uncheck all with a double click on the checkbox column (S)
-new function (right click with classification preview pointer) displays the algorithm raster of the preview (a raster where every pixel has the value calculated by the algorithm for the class assigned to the pixel thereof) useful for assessing spectral signatures (black pixels are distant from the spectral signatures and white pixels are closer to the spectral signatures)
-now a classification style (.qml file) is saved along with the classification output allowing for displaying classification colors when loading the classification in new qgis projects
-new buttons in tab Band set for exporting the band set to virtual raster or stacked raster (a multi-band .tif file)
-now the signature file path is stored in the project as absolute or relative path according to QGIS project settings
-in Settings added an option for the creation of virtual rasters for certain temporary data (which saves disk space)
-new button in Debug for testing SCP dependencies
-enhanced the Maximum Likelihood algorithm which is now considerably faster
-temporary ROI are displayed as overlay (rubber band) with a + symbol on clicked pixel
-new option to hide and show the temporary ROIs
-in ROI dock new option for refreshing automatically the temporary ROI while changing ROI parameters
-in ROI dock new option for automatic plot of temporary ROI
-new option for displaying over the region growing cursor vegetation indices such as NDVI and EVI (the latter only for images converted to reflectance)
-new function for displaying the pixel spectral signature with right click on map using ROI pointer
-new combo box in SCP toolbar for quickly displaying RGB color composites of input image (and automatic creation of a virtual raster for band set)
-now the window spectral signature plot allows for the display of signature values in table
-new tool Split in Pre processing tab for splitting raster bands
-new tool Convert classification to vector in Post processing tab for creating a classification shapefile
-new tool Reclassification in Post processing tab for the easy reclassification from Class ID to Macroclass ID and Python expressions are allowed in the definition (e.g. raster < 3)
-enhanced function Clip multiple rasters
-new reset button for resetting the signature list file
-new button in the classification dock for merging spectral signatures obtaining a new one calculated as the average of signature values (without covariance matrix)
-now classes or macroclasses having ID = 0 are allowed and classified as 0 (unclassified) in the output which is useful for creating masks for classes
-in the signature list it is possible to set the color of multiple selected signatures at once (pressing shift and double click on a color) and check/uncheck all with a double click on the checkbox column (S)
-new function (right click with classification preview pointer) displays the algorithm raster of the preview (a raster where every pixel has the value calculated by the algorithm for the class assigned to the pixel thereof) useful for assessing spectral signatures (black pixels are distant from the spectral signatures and white pixels are closer to the spectral signatures)
-now a classification style (.qml file) is saved along with the classification output allowing for displaying classification colors when loading the classification in new qgis projects
-new buttons in tab Band set for exporting the band set to virtual raster or stacked raster (a multi-band .tif file)
-now the signature file path is stored in the project as absolute or relative path according to QGIS project settings
-in Settings added an option for the creation of virtual rasters for certain temporary data (which saves disk space)
-new button in Debug for testing SCP dependencies
Possibly, other functions could be added to the plugin, and I hope to update the user manual before the release date.
Soon, a beta version of SCP 4.0 will be available for the users of the Semi-Automatic Classification Plugin through the Facebook group and the Google+ Community.
